CN112929605B - Intelligent PTZ camera cruising method considering semantics - Google Patents
Intelligent PTZ camera cruising method considering semantics Download PDFInfo
- Publication number
- CN112929605B CN112929605B CN202110115187.XA CN202110115187A CN112929605B CN 112929605 B CN112929605 B CN 112929605B CN 202110115187 A CN202110115187 A CN 202110115187A CN 112929605 B CN112929605 B CN 112929605B
- Authority
- CN
- China
- Prior art keywords
- cruising
- grid
- monitoring area
- camera
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/20—Checking timed patrols, e.g. of watchman
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/695—Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a PTZ camera intelligent cruise method considering semantics. The method comprises the following steps: (1) determining an effective monitoring area according to the three-dimensional geographic information and the vision field boundary; (2) performing regular cruise on the effective monitoring area to obtain a video frame image sequence, splicing the video frame image sequence into a wide-view-angle large image based on a characteristic point matching algorithm, and calculating a mapping matrix from the large image to a geographic space; (3) performing real-time cruising according to a regular cruising method, updating a large image by using a real-time video frame image, and mapping a target detection result to a geographic space; (4) carrying out grid division on the effective monitoring area, counting the number of targets in each grid, and determining a hotspot set; (5) determining an optimal cruising path through an exhaustion method; (6) cruising according to the optimal path, updating a large map, target detection and map mapping in real time, and continuously cruising for a dt time period; (7) and (5) repeating the steps (2) to (6). The method solves the problem that the PTZ camera is easy to lose important information in scene monitoring.
Description
Technical Field
The invention relates to a PTZ camera intelligent cruising method considering semantics, and belongs to the technical field of scientific cross fusion of videos and geographic information.
Background
With the rapid development of information technology, video monitoring technology plays an increasingly important role in smart cities, emergency disaster relief, natural resource monitoring and other fields. The monitoring video has the advantages of high definition, reality and real time, the geographic information system has unique spatial analysis capability and can predict and simulate the development trend of things, the real-time monitoring information of a geographic scene can be provided through the cross fusion of the two, and the interaction between people and the geographic scene is realized. The video images collected by the camera are intelligently analyzed by combining geographic semantics and space-time hot spots, so that the reliability of the monitoring video is improved.
At present, the monitoring of scene on a large scale adopts the PTZ camera more, and for the static camera that the sight field is restricted, the PTZ camera has more nimble monitoring mode, can catch the target rapidly, realizes the continuous effectual control to whole control area. However, in practical applications, the PTZ camera easily ignores surrounding actual scenes, monitors invalid areas, and causes loss of important information in the monitored areas. If the monitoring area can be determined by considering geographic semantics, the empty hot spots monitor the monitoring area in a block grading manner when in combination in geographic space, and cruise is carried out according to preset points, so that the cruise time of each block area can be properly prolonged or shortened by the camera, the cruise efficiency can be greatly improved, and more valuable information can be recorded. At present, the research field mainly focuses on the intelligent analysis of PTZ camera video images, and the PTZ camera video images lack close combination with geographic information and cannot be monitored in consideration of the geographic information.
Disclosure of Invention
The invention provides a PTZ camera intelligent cruise method considering semantics to solve the problem that important information is easily lost in scene monitoring of a PTZ camera, so that the cruise quality is improved.
The technical idea of the invention is as follows: determining an effective monitoring area in consideration of geographical semantics; splicing a video frame image sequence of a PTZ complete monitoring area based on SIFT feature point matching, and updating and splicing a wide-view-angle large image by real-time video matching; carrying out grid division on an effective monitoring area in a geographic space, and simultaneously carrying out target detection and geographic mapping on a real-time video; and analyzing the monitoring area and the target information thereof by combining the GIS to obtain an optimal cruising path which passes through all hot spots and has the minimum total cruising distance so as to realize intelligent cruising of the PTZ camera.
The technical scheme adopted by the invention is as follows:
a PTZ camera intelligent cruise method considering semantics comprises the following steps:
and 7, repeating the steps 2-6 after the optimal cruise path is cruising for the dt time period.
Further, in step 2, the specific method for performing regular cruise on the effective monitoring area is as follows: the active monitoring area is navigated in a "Z" shape in the order from top to bottom, left to right.
Further, in the step 2, when the image sequence is spliced into the wide-view-angle large image P, a certain degree of overlapping between the images is required.
Further, in the step 2, when the mapping matrix H from the wide-view-angle large graph P to the geographic space is solved, the number of the selected homonymous points is greater than or equal to 4.
Further, in step 3, when detecting the target in the real-time video, a time threshold T is setminWith TminAs time intervals, for TminDetecting the target in the time period and simultaneously recording the current TminNumber of targets and location information within a time period.
Further, in step 4, when regular grid division is performed on the effective monitoring area in the geographic space, the effective monitoring area is divided by m × n square grids, and the center coordinate of each square grid represents the position of each grid.
Further, in step 4, when determining the hot spot set, the level of each grid is divided by a natural breakpoint method according to the number of targets in each grid, so as to represent the target aggregation degree of each grid, the grid center with target aggregation is marked as a hot spot, the grid center without target is marked as a non-hot spot, and the more the number of targets in the grid is, the longer the camera stays during the cruise.
Further, in the step 6, the grid center with the most aggregated target number is used as an initial hot spot, when the camera passes through each hot spot, the attitude of the camera is adjusted according to the calculated PTZ value of each hot spot, and the camera returns to the initial hot spot after continuously cruising dt time period.
Compared with the prior art, the method of the invention has the following technical effects:
(1) the invention considers that the current PTZ camera does not consider geographical semantics during monitoring, and easily monitors a region with few or no targets during monitoring.
(2) According to the invention, the video frame image sequence of the monitored scene is spliced into the wide-view-angle large image, the large image is updated by matching the real-time video frame images, and the hot spot set in the geographic space is subjected to hierarchical cruising, so that the real-time performance and the applicability of the PTZ camera in cruising of an important monitored area are ensured.
(3) The invention combines the time-space hot spots, uses the regular grids to divide the effective monitoring area in the geographic space, extracts the target information in each grid, takes the target aggregation area as the cruise key area, properly prolongs or shortens the cruise time of each area according to the aggregation degree of the target, ensures the cruise flexibility of the camera, and realizes the intelligent cruise taking the time-space hot spots into consideration.
(4) Aiming at the hot spot information, the optimal cruise path which passes through all hot spot sets and has the minimum total cruise distance is determined, and the PTZ value is determined according to the cruise hot spots and the camera center so as to adjust the posture of the camera passing through each hot spot, so that the camera center is over against each hot spot, the intelligent cruise efficiency of the PTZ camera is ensured, and the cruise efficiency is improved.
Drawings
FIG. 1 is a block flow diagram of the method of the present invention;
FIG. 2 is a schematic illustration of determining an actual active surveillance zone for a PTZ camera;
FIG. 3 is a schematic illustration of a mesh dissecting a monitored area;
FIG. 4 is a schematic illustration of determining an optimal cruise path.
Detailed Description
The following description will explain embodiments of the present invention in further detail with reference to the accompanying drawings.
The PTZ camera intelligent cruise method considering the semantics comprises the following steps: firstly, determining an effective monitoring area V of the PTZ camera according to the center of the camera, the vision field boundary (horizontal and vertical) and three-dimensional fine geographic information (ground objects, a building distribution diagram and DEM data thereof); cruising the effective monitoring area V regularly to obtain a video frame image sequence of the PTZ monitoring area, matching and splicing the video frame image sequence into a wide-view-angle large image P based on SIFT feature points, and simultaneously resolving a mapping matrix H from the image P to a geographic space according to a fine map or a remote sensing image; secondly, cruising is carried out according to a regular cruising method, the wide-view-angle large image P is updated through real-time video frame image matching acquired through cruising, target detection is carried out, and a target detection result is mapped to a geographic space; subdividing an effective monitoring area V under a geographic space by using m multiplied by n square grids, and counting TminNumber and location information of targets in each grid within a time periodDetermining a hotspot set, recording the grid centers with target aggregation as hotspots, and recording the grid centers without targets as non-hotspots; determining a path which passes through each hot point and has the minimum total cruising distance as an optimal cruising path according to the hot point set; and finally, updating the large map, the target detection and the map mapping in real time according to the optimal cruise Path, meanwhile, calculating a corresponding PTZ value according to the coordinate of the center point of the camera and the hot point coordinate of the optimal cruise Path Path, adjusting the posture of the camera passing through each hot point, and continuously cruising for a dt time period.
In a specific practical application process, the method of the embodiment specifically comprises the following steps for a single PTZ camera video:
And 2, acquiring a video frame image sequence of the PTZ complete monitoring area according to a regular cruising mode, splicing a plurality of frames of video images into a wide-view-angle large image P based on SIFT feature point detection matching, wherein a certain overlapping degree exists between the video frame images. In this embodiment, the active monitoring area is navigated in a "Z" shape from top to bottom and left to right. Aiming at a monitoring video of a PTZ camera, selecting more than 4 homonymous points by combining a corresponding high-definition remote sensing map or a three-dimensional fine map, and resolving a homography matrix from a wide-view-angle large map P to a geographic space; let a geographic coordinate point M (X, Y), a pixel coordinate M (X, Y) of the image P, and a mapping matrix between the two points be H, i.e.
When selecting the same-name point, look at firstAnd selecting the frequency image, selecting corresponding points on the high-definition remote sensing base map, and resolving by selecting 4 pairs or more.
And 3, performing real-time cruising on the effective monitoring area according to the Z-shaped regular cruising method, updating the image P by using a real-time video frame image, and mapping the target detection result of the real-time video to the geographic space.
Firstly, according to the homonymous feature points matched with the SIFT feature points, obtaining the mapping relation from each frame of video image to the wide-view-angle large image P, and setting the pixel coordinate m (x, y) of the image P and the pixel coordinate m '(x', y ') of the real-time video frame image, wherein the mapping matrix between the two points is H', namely
Then, based on the mapping matrix H', fusing the real-time video content to a wide-view-angle large map P to obtain a real-time video map; and finally, detecting targets in the real-time video, such as pedestrians, vehicles and the like by taking an SSD (Single Shot MultiBox Detector) model as a target detection model, and mapping the real-time video frame image to the geographic space based on the target detection result based on the homography matrixes H and H'. When detecting a target in a real-time video, setting a time threshold TminWith TminAs time intervals, for TminDetecting the target in the time period and simultaneously recording the current TminNumber of targets and location information within a time period.
And 4, carrying out grid division on the effective monitoring area of the PTZ camera, counting the number and the position information of targets in each grid, and determining a hotspot set.
Firstly, longitude and latitude coordinates (if precision requirement is high, the longitude and latitude coordinates can be changed into projected plane rectangular coordinates) and grid size of a lower left corner and an upper right corner of a rectangle formed by the whole grid are obtained, then the column number n is calculated by dividing the difference between the longitude of the upper right corner and the lower left corner by the grid size, and the row number m is calculated by dividing the difference between the latitude of the upper right corner and the lower left corner by the grid size. And determining the geographic coordinates of the central points of the grids according to the longitude and latitude coordinates of the lower left corner and the upper right corner and a certain rule. If the grid starts from 0 row and 0 column from left to right and from bottom to top, and has m-1 rows and n-1 columns, then
The longitude of the center point of each grid is equal to the longitude of the lower right corner + (0.5+ the number of columns of the grid) and the grid size,
the latitude of the center point of each grid is equal to the lower right corner latitude + (0.5+ the number of rows of the grid) and the grid size;
the specific implementation of the grid generation method can adopt an ArcGIS fishing net (fishernet) tool, and the method can also be realized from the bottom layer, and fig. 3 is a grid subdivision schematic diagram of an effective monitoring area of a PTZ camera. Then, each grid is used as an area, the coordinates of the central point of each grid are used as the position center of the area, the number and the position information of the targets in each grid are counted, each grid with target aggregation is used as a hot spot set, the grid center with target aggregation is marked as a hot spot, and the grid center without target aggregation is a non-hot spot. In the hotspot set, the more the number of the targets is, the higher the target aggregation degree of the area relative to other areas is, and the higher the importance degree in the monitoring scene is; conversely, the lower. Finally, according to the target aggregation degrees of different hot point sets, the cruising time of the camera in the area is properly prolonged or shortened, and the higher the aggregation degree of the target is, the longer the cruising time of the camera in the area is; conversely, it becomes shorter.
And 5, determining a Path which passes through all hot points and has the minimum total cruising distance as an optimal cruising Path through an exhaustion method, and calculating a corresponding PTZ value according to the coordinate of the center point of the camera and the coordinate of the hot point of the optimal cruising Path.
Firstly, regarding w hot spot sets, grid labels are respectively carried out according to a Z shape, namely from top to bottom and from left to right, and the total number of the grids is w. Then by exhaustion method, with the most concentrated grid centers w of the target1As the initial hotspot, connect each hotspot to obtain (w-1)! A strip path; and finally, finding a Path which passes through the w hot points and has the minimum total cruising distance as an optimal cruising Path.
As shown in FIG. 4, the grids are numbered from top to bottom, from left to right, to obtain 13 hot spot grids, and a total of (13-1)! The seed path can be selected, and the closed path of 4-5-6-7-8-9-10-11-12-13-1-2-3-4 is selected as the optimal cruising path.
And 6, cruising according to the optimal Path, updating the wide-view-angle large map in real time, carrying out target detection and map mapping, and meanwhile, calculating the corresponding PTZ value of each hot spot according to the coordinate of the center point of the camera and the hot spot coordinate of the optimal cruising Path Path.
Let geographical coordinates (x) of the center point of the camera1,y1,z1) Cruise Path Path a certain hot spot P1Has the coordinates of (x)2,y20), then there are:
azimuth angle Pan:
Pan＝arcsinPan
the inclination angle Tilt:
Tilt＝arccosTilt
and the Z value of the camera can be subjected to zoom adjustment according to requirements of imaging size, image definition and the like after the azimuth angle Pan and the inclination Tilt are determined.
And taking the grid center with the largest target aggregation number as an initial hot point, adjusting the optimal cruise attitude of the camera according to the calculated PTZ value when passing through each hot point set, enabling the center of the camera to be over against the hot point, properly prolonging or shortening the cruise time according to the aggregation degree of the targets in each hot point set, and continuing to cruise for a dt time period. And returning to the starting hot spot after the optimal path cruise dt time period, and repeating the steps 2-6 to realize the intelligent cruise of the PTZ camera.
Claims (8)
1. A PTZ camera intelligent cruise method considering semantics is characterized by comprising the following steps:
step 1, determining an effective monitoring area according to the center of a camera, a view field boundary and three-dimensional fine geographic information;
step 2, performing regular cruise on the effective monitoring area to obtain a video frame image sequence of the complete monitoring area of the PTZ camera, splicing the image sequence into a wide-view-angle large image P based on SIFT feature point matching, and resolving a mapping matrix H from the wide-view-angle large image P to a geographic space according to a fine map or a remote sensing image;
step 3, performing real-time cruising on the effective monitoring area according to a regular cruising method, and updating the wide-view-angle large image P by matching real-time video frame images; then, detecting a target in the real-time video, and mapping a target detection result to a geographic space;
step 4, regular grid division is carried out on the effective monitoring area in the geographic space, the target number and the position information in each grid are counted, and a hotspot set is determined; the more the number of the targets in each grid is, the higher the aggregation degree of the targets in the area where the grid is located is, and the longer the time for the camera to cruise the area corresponding to the grid is; otherwise, the time is shortened;
step 5, determining a path which passes through all hot spots and has the minimum total cruising distance as an optimal cruising path by an exhaustion method;
step 6, cruising is carried out on the effective monitoring area according to the optimal cruising path, a wide-view-angle large map P, target detection and map mapping are updated in real time, and meanwhile, a corresponding PTZ value is calculated according to the central point coordinate of the PTZ camera and the hot point coordinate of the optimal cruising path; adjusting the posture of the camera passing through each hot spot, and continuously cruising for a dt time period;
and 7, repeating the steps 2-6 after the optimal cruise path is cruising for the dt time period.
2. The PTZ camera intelligent cruise method considering semantics as claimed in claim 1, wherein in the step 2, a specific method for performing regular cruise on an effective monitoring area is as follows: the active monitoring area is navigated in a "Z" shape in the order from top to bottom, left to right.
3. The PTZ camera intelligent cruising method considering semantics as claimed in claim 1, wherein in the step 2, when the image sequence is spliced into a large map P with a wide viewing angle, a certain degree of overlapping between the images is required.
4. The PTZ camera intelligent cruising method considering semantics as claimed in claim 1, wherein in the step 2, when a mapping matrix H from a wide view angle large graph P to a geographic space is solved, the number of selected points with the same name is greater than or equal to 4.
5. The PTZ camera intelligent cruising method considering semantics as claimed in claim 1, wherein in the step 3, a time threshold T is set when detecting the target in the real-time videominIn terms of TminAs time intervals, for TminDetecting the target in the time period and simultaneously recording the current TminNumber of targets and location information within a time period.
6. The PTZ camera intelligent cruising method considering semantics as claimed in claim 1, wherein in the step 4, when regular grid division is performed on an effective monitoring area under a geographic space, the effective monitoring area is divided by m × n square grids, and a central coordinate of each square grid represents a position of each grid.
7. The PTZ camera intelligent cruising method considering semantics as claimed in claim 1, wherein in the step 4, when determining the hot spot set, the grades of the grids are divided by a natural breakpoint method according to the number of targets in the grids so as to represent the target aggregation degree of the grids, and the grid centers with target aggregation are marked as hot spots, and the grid centers without targets are marked as non-hot spots.
8. The PTZ camera intelligent cruising method considering semantics as claimed in claim 1, wherein in the step 6, a mesh center with the largest number of target aggregation is used as an initial hot spot, when the camera passes through each hot spot, the attitude of the camera is adjusted according to the calculated PTZ value of each hot spot, and after the camera continues to cruise for a dt time period, the camera returns to the initial hot spot.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110115187.XA CN112929605B (en) | 2021-01-28 | 2021-01-28 | Intelligent PTZ camera cruising method considering semantics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110115187.XA CN112929605B (en) | 2021-01-28 | 2021-01-28 | Intelligent PTZ camera cruising method considering semantics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112929605A CN112929605A (en) | 2021-06-08 |
| CN112929605B true CN112929605B (en) | 2022-06-24 |
Family
ID=76167449
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110115187.XA Active CN112929605B (en) | 2021-01-28 | 2021-01-28 | Intelligent PTZ camera cruising method considering semantics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112929605B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101832779A (en) * | 2010-03-03 | 2010-09-15 | 北京大学 | Navigation method in complicated environment |
| CN103576683A (en) * | 2012-08-03 | 2014-02-12 | 中国科学院深圳先进技术研究院 | Scheduling method and system for multiple patrol robots |
| JP2016147556A (en) * | 2015-02-10 | 2016-08-18 | 株式会社デンソー | Drive control device and drive control method |
| CN106506955A (en) * | 2016-11-10 | 2017-03-15 | 国网江苏省电力公司南京供电公司 | A kind of transformer substation video polling path planing method based on GIS map |
| CN107360394A (en) * | 2017-06-16 | 2017-11-17 | 河北汉光重工有限责任公司 | More preset point dynamic and intelligent monitoring methods applied to frontier defense video monitoring system |
| CN109358650A (en) * | 2018-12-14 | 2019-02-19 | 国网冀北电力有限公司检修分公司 | Polling path planing method, device, unmanned plane and computer readable storage medium |
| CN110163990A (en) * | 2019-05-24 | 2019-08-23 | 中铝视拓智能科技有限公司 | A kind of mobile robot method for inspecting and system |
| CN110991272A (en) * | 2019-11-18 | 2020-04-10 | 东北大学 | Multi-target vehicle track identification method based on video tracking |
| CN111836002A (en) * | 2019-04-15 | 2020-10-27 | 浙江宇视科技有限公司 | Video polling method, device and server |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102685460A (en) * | 2012-05-17 | 2012-09-19 | 武汉大学 | Video monitoring and cruising method for integrating measurable scene image and electronic map |
-
2021
- 2021-01-28 CN CN202110115187.XA patent/CN112929605B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101832779A (en) * | 2010-03-03 | 2010-09-15 | 北京大学 | Navigation method in complicated environment |
| CN103576683A (en) * | 2012-08-03 | 2014-02-12 | 中国科学院深圳先进技术研究院 | Scheduling method and system for multiple patrol robots |
| JP2016147556A (en) * | 2015-02-10 | 2016-08-18 | 株式会社デンソー | Drive control device and drive control method |
| CN106506955A (en) * | 2016-11-10 | 2017-03-15 | 国网江苏省电力公司南京供电公司 | A kind of transformer substation video polling path planing method based on GIS map |
| CN107360394A (en) * | 2017-06-16 | 2017-11-17 | 河北汉光重工有限责任公司 | More preset point dynamic and intelligent monitoring methods applied to frontier defense video monitoring system |
| CN109358650A (en) * | 2018-12-14 | 2019-02-19 | 国网冀北电力有限公司检修分公司 | Polling path planing method, device, unmanned plane and computer readable storage medium |
| CN111836002A (en) * | 2019-04-15 | 2020-10-27 | 浙江宇视科技有限公司 | Video polling method, device and server |
| CN110163990A (en) * | 2019-05-24 | 2019-08-23 | 中铝视拓智能科技有限公司 | A kind of mobile robot method for inspecting and system |
| CN110991272A (en) * | 2019-11-18 | 2020-04-10 | 东北大学 | Multi-target vehicle track identification method based on video tracking |
Non-Patent Citations (2)
| Title |
|---|
| 巡检线路排班的优化模型研究;陈辉;《四川理工学院学报(自然科学版)》;20180220(第01期);全文 * |
| 视频和GIS协同的人群状态感知模型;邓仕虎等;《信阳师范学院学报(自然科学版)》;20180131(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112929605A (en) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10630899B2 (en) | Imaging system for immersive surveillance | |
| CN110674746B (en) | Method and device for realizing high-precision cross-mirror tracking by using video spatial relationship assistance, computer equipment and storage medium | |
| US8564643B1 (en) | Camera image georeferencing systems | |
| US6741725B2 (en) | Motion tracking using image-texture templates | |
| CN110516014B (en) | Method for mapping urban road monitoring video to two-dimensional map | |
| SG191198A1 (en) | Imaging system for immersive surveillance | |
| WO2021077607A1 (en) | Data processing method and apparatus | |
| US20120027371A1 (en) | Video summarization using video frames from different perspectives | |
| EP1877987A1 (en) | Automated surveillance system | |
| CN111143504B (en) | Multi-camera indoor video map construction method | |
| CN112040265B (en) | Multi-camera collaborative geographic video live broadcast stream generation method | |
| US11588975B2 (en) | Method of mapping monitoring point in CCTV video for video surveillance system | |
| CN115375779B (en) | Method and system for camera AR live-action annotation | |
| CN111683221B (en) | Real-time video monitoring method and system for natural resources embedded with vector red line data | |
| US11703820B2 (en) | Monitoring management and control system based on panoramic big data | |
| CN113869231A (en) | Method and equipment for acquiring real-time image information of target object | |
| CN109712249B (en) | Geographic element augmented reality method and device | |
| CN113192125B (en) | Multi-camera video concentration method and system in virtual viewpoint-optimized geographic scene | |
| CN112929605B (en) | Intelligent PTZ camera cruising method considering semantics | |
| CN114066945B (en) | Video tracking method and system based on pixel spatial resolution | |
| CN106777172A (en) | A kind of video Space-time Search method | |
| CN112114301A (en) | Target positioning method, device, computer readable storage medium and computer equipment | |
| Janoos et al. | Sensor configuration for coverage optimization in surveillance applications | |
| EP4365839A1 (en) | A computer implemented method for determining 3d coordinates describing a scene, a computer program, a computer program product, a device for determining 3d coordinates describing a scene and a display device for display of information relating to 3d coordinates describing a scene | |
| CN118379675A (en) | Trampling risk early warning method and device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |